Lightspeed's Connor Love on drone warfare, counter-UAS investing, and why the US is dangerously behind

Summary

Connor Love, a partner at Lightspeed Venture Partners and former US military officer, argues that Ukraine's weekend drone strike on Russian bases — what he calls a "Trojan horse" moment — is the clearest signal yet of where warfare is heading: cheap, attritable, autonomous systems at scale.

The attack, which Love describes as a flawless execution based on open-source intelligence, used FPV drones delivered via civilian trucks and guided over Russian cellular networks. The cellular dependency is real, but shutting down the network is not a clean countermeasure — much Russian military coordination runs over WhatsApp, so killing the cell network degrades the defender's own response capability. Love's broader point is that layered defense is the only viable architecture. No single system — electronic warfare jamming, laser-based intercepts, kinetic interceptors like Anduril's Anvil drone, or better counterintelligence — solves the problem alone.

The US manufacturing gap

The production numbers expose the core problem. The leading US drone manufacturer in its class is producing roughly 1,500 units per month. Ukraine is building thousands per day. Love frames this as a structural gap that procurement reform alone cannot close — supply chains, tooling, and training pipelines take years to build. He says he spent a conversation with a member of Congress that morning on exactly this: even if acquisition pathways are fixed tomorrow, the physical capacity to field systems at scale does not exist yet.

The culture inside DoD is shifting, if slowly. Giving $100 million upfront to a company like Anduril or Castilian ahead of delivery is starting to be seen as the right answer over routing money through traditional primes — but Love is direct that the US is still far from that norm.

The China lens

Love's read is that Ukraine is a rehearsal for the Pacific. Chinese military planners are watching the Russian losses and updating both their offensive playbook and their defensive posture. A Chinese official recently posted a render of a mothership aircraft launching hundreds of drones simultaneously — aspirational, but directionally telling. The more uncomfortable point Love makes is doctrinal: the US keeps humans in the loop on nearly every lethal decision, as a matter of law and values. China, he argues, will not apply that constraint in a conflict, and the asymmetry matters.

DJI and the dual-use supply chain

The DJI ban debate is more concrete than the TikTok fight ever was. Love notes that early in his military career, a team handed his unit a DJI drone out of a box as their introduction to drone warfare. The data and feed risks are real, but the deeper concern is industrial: Chinese consumer drone demand has built a manufacturing base and supply chain with genuine dual-use capacity, regardless of whether any individual product reaching the US is weaponized.

Counter-UAS market structure

Love maps the counter-UAS market as a spectrum. At one end, non-kinetic systems — EW jamming, lasers. At the other, kinetic options from shotguns to interceptor drones. In the middle, hybrid approaches combining EW-resilient hardware with collision-based intercepts. He singles out a layer that gets less attention: the intelligence and detection layer that ideally prevents an attack from being staged in the first place. He also flags Cape, founded by early Palantir veteran John Doyle, as worth watching — it is building a private cellular network designed to keep communications alive when public networks are shut down or compromised.

Golden Dome

On Golden Dome, Love's honest assessment is that nobody inside or outside government knows yet what it actually is. He credits Trump for using ambitious goal-setting to align congressional and DoD support — calling it a "put the cart before the horse" approach that is sometimes the only way to get large programs moving. The substantive case is straightforward: Israel has a layered missile defense system that the US helped develop, and the US has no domestic equivalent. Building one will require a Manhattan Project-style coordination across space, missile defense, and new budget lines. Love thinks the real answer to mass drone and missile saturation is not more $10 million interceptor missiles but cheaper, high-performance alternatives that change the cost calculus for attackers.

Nominal and the picks-and-shovels layer

Love is a backer of Nominal, which builds telemetry and software infrastructure for hardware development programs. His thesis is that the shift toward SpaceX-style iterative hardware development — build fast, test in flight — creates demand for a software layer that traditional defense programs never needed. Getting to a million drones in the hands of warfighters requires supply chain software, telemetry, and development tooling below the hardware layer. He sees Nominal as one example of a picks-and-shovels opportunity that will produce winners alongside the hardware primes.